6307 2888 weka.FilteredClassifier_HoeffdingTree weka.classifiers.meta.FilteredClassifier 1 Weka_3.8.1_12647 Weka implementation of FilteredClassifier 2017-04-17T15:23:23 English Weka_3.8.1 -do-not-check-capabilities flag If set, classifier capabilities are not checked before classifier is built (use with caution). -doNotMakeSplitPointActualValue flag Do not make split point actual value. A flag Laplace smoothing for predicted probabilities. B flag Use binary splits only. C option Set confidence threshold for pruning. (default 0.25) D flag Output binary attributes for discretized attributes. E flag Use better encoding of split point for MDL. F option weka.filters.supervised.attribute.Discretize -R first-last -precision 6 Full class name of filter to use, followed by filter options. eg: "weka.filters.unsupervised.attribute.Remove -V -R 1,2" J flag Do not use MDL correction for info gain on numeric attributes. K flag Use Kononenko's MDL criterion. L flag Do not clean up after the tree has been built. M option Set minimum number of instances per leaf. (default 2) N option Set number of folds for reduced error pruning. One fold is used as pruning set. (default 3) O flag Do not collapse tree. Q option Seed for random data shuffling (default 1). R flag Use reduced error pruning. S flag Do not perform subtree raising. U flag Use unpruned tree. V flag Invert matching sense of column indexes. W baselearner weka.classifiers.trees.HoeffdingTree Full name of base classifier. (default: weka.classifiers.trees.J48) Y flag Use bin numbers rather than ranges for discretized attributes. batch-size option The desired batch size for batch prediction (default 100). num-decimal-places option The number of decimal places for the output of numbers in the model (default 2). output-debug-info flag If set, classifier is run in debug mode and may output additional info to the console precision option Precision for bin boundary labels. (default = 6 decimal places). W 5943 2747 weka.HoeffdingTree weka.classifiers.trees.HoeffdingTree 6 Weka_3.8.1_11006 Geoff Hulten, Laurie Spencer, Pedro Domingos: Mining time-changing data streams. In: ACM SIGKDD Intl. Conf. on Knowledge Discovery and Data Mining, 97-106, 2001. 2017-03-30T15:20:36 English Weka_3.8.1 E option 1.0E-7 The allowable error in a split decision - values closer to zero will take longer to decide (default = 1e-7) G option 200.0 Grace period - the number of instances a leaf should observe between split attempts (default = 200) H option 0.05 Threshold below which a split will be forced to break ties (default = 0.05) L option 2 The leaf prediction strategy to use. 0 = majority class, 1 = naive Bayes, 2 = naive Bayes adaptive. (default = 2) M option 0.01 Minimum fraction of weight required down at least two branches for info gain splitting (default = 0.01) N option 0.0 The number of instances (weight) a leaf should observe before allowing naive Bayes to make predictions (NB or NB adaptive only) (default = 0) P flag Print leaf models when using naive Bayes at the leaves. S option 1 The splitting criterion to use. 0 = Gini, 1 = Info gain (default = 1) Verified_Supervised_Classification weka weka_3.8.1 https://api.openml.org/data/download/4699324/weka.classifiers.trees.HoeffdingTree641263798600590420.class class e13ea9ef1e91673a0473e7831a351c34 Verified_Supervised_Classification weka weka_3.8.1 https://api.openml.org/data/download/5035525/weka.classifiers.meta.FilteredClassifier511884703049568570.class class d36a54655368c6b3970f0bd7a40dda83